This experiment investigated participants' ability to keep track of time during a visual and memory search task where task difficulty and duration were manipulated. Two hundred and ninety-two participants performed the task for eight or 58 minutes. Participants in the prospective time judgment condition were forewarned of an impending time estimate whereas those in the retrospective condition were not. Cognitive load was manipulated and assessed by assigning participants to either a consistent or a varied mapping condition. The results revealed overestimation and higher variability of estimates in the prospective condition compared to the retrospective one in the eight-minute task only. Moreover, participants significantly overestimated the duration of the eight-minute task and underestimated the 58-minute task. Finally, cognitive load had no effect on participants' time estimates. Thus, the well-known cross-over interaction between cognitive load and estimation paradigm (Block et al., 2010) does not seem to extend to longer durations.